TapCub is live — analytics, insights and live chat, free on one platform
Company

See the data clearly. Find the growth.

TapCub is one product for website analytics, product insights and live chat — built by a small team that got tired of stitching three tools together and arguing about whose numbers were right.

Mission

See the data clearly, find the growth.

Most teams do not lack data. They lack numbers they trust, in one place, with the next action attached.

A marketing lead opens three dashboards and gets three different visitor counts. A product manager builds a funnel in one tool and cannot see the chat that happened at step three. A support agent answers a question without knowing which campaign brought the person in. We built TapCub to close those gaps: one snippet, one visitor model, one place where the count, the analysis and the conversation line up.

That is what the name of this page means. "See the data clearly" is about honest counts — humans separated from bots, estimates labelled as estimates. "Find the growth" is about what you do next: the drop in a funnel, the channel that pays back, the visitor worth talking to right now.

What we believe

Four things we will not trade away

They shape every feature decision, including the ones we say no to.

Real numbers

Humans and bots stay on separate lines. AI crawlers get their own ledger. When a number is an estimate — unique visitors without cookies, for example — the product says so instead of pretending.

Privacy by default

The analytics script sets no cookies and raw IP addresses never touch the database. Chat writes a single session token only after the visitor speaks. Compliance is a setting, not a banner.

Simple to use

One line of code, first numbers within a minute, reports that answer a question rather than offer forty filters. If a feature needs a training session, we have not finished it.

Analytics, insights and chat as one

The same visitor model powers the pageview count, the retention grid and the chat inbox. You do not export from one tool to import into another; you click through.

Product principles

How we decide what to build

Six rules that come up in almost every design discussion.

  1. 1

    Answer the question, then show the chart

    Every report starts with the sentence a person would say in a meeting — "sign-ups fell 18% because checkout slowed down" — and the chart proves it.

  2. 2

    Count honestly, even when it looks worse

    Filtering bots usually lowers the headline number. We show the lower number by default and let you switch to "all traffic" when you need it.

  3. 3

    Make the private option the easy option

    Cookieless is the default. Identification mode, longer retention and profile fields are opt-in with the consequences spelled out in the compliance centre.

  4. 4

    One model for web, app and mini-program

    A person who installs your app after reading a blog post is one person. Seven SDKs report into the same event store with the same identity rules.

  5. 5

    Templates over blank pages

    Ten industry templates install the metrics, dashboards and funnels a sector actually uses, so the first useful report exists before you write a tracking plan.

  6. 6

    Ship in the open

    Every change lands on the public changelog with the date. If something regresses, the entry says so.

In numbers

Facts you can check in the product

No customer counts, no funding rounds — just the figures the product is built around.

analytics script, gzipped
≈2 KB
analytics script, gzipped
platforms with an SDK
7
platforms with an SDK
ad platforms with click IDs
14
ad platforms with click IDs
industry templates
10
industry templates
bot and crawler rules
197+
bot and crawler rules

Figures are maintained as the product evolves; the console and documentation are authoritative for the current values.

How we got here

The product, stage by stage

Written from the capabilities as they shipped, not from a pitch deck.

  1. Stage 1

    A web analytics counter that filtered bots

    TapCub started as a plain website counter with one obsession: separating humans from crawlers before the number reached a dashboard. Realtime, sources, pages and a five-layer bot filter came first.

    Website analytics

  2. Stage 2

    Live chat on the same snippet

    Instead of a second script, chat loaded from the analytics snippet and read the same visit context. Inbox, agent roles, proactive invites, two-way translation and AI replies followed, each with its own privacy review.

    Live chat

  3. Stage 3

    Two products, two subscriptions

    Analytics and chat became separate plans with their own free tiers, so a blog does not pay for agents and a shop does not pay for funnels it never opens.

    Pricing

  4. Stage 4

    Behavioural analytics and seven SDKs

    Identity rules, an event store, funnels, retention, paths and heatmaps arrived together with SDKs for iOS, Android, HarmonyOS, Flutter, React Native, uni-app and mini-programs. The AI event inbox began naming unplanned events.

    Product insights

  5. Stage 5

    Marketing attribution and industry templates

    Click IDs from 14 ad platforms, first- and last-touch attribution, channel ROI against imported spend, and ten installable industry templates that set up metrics and dashboards per sector.

    Industry templates

  6. Stage 6

    Compliance centre, payments and the docs site

    Cookieless presets, consent and deletion workflows, card and wallet payments through Stripe, PayPal, Alipay and WeChat Pay, multi-factor sign-in and a bilingual documentation site — the pieces a team needs before it trusts a tool with production traffic.

    Privacy and security

  7. Now

    This website, and the next changelog entry

    The site you are reading was rebuilt around the same honesty rule as the product: sample data is labelled, comparisons are by category, and the changelog is public.

    Changelog

How we work

A small, remote team with written habits

Three practices that explain most of how the product feels.

Remote, across time zones

We work from several cities and overlap for a few hours a day. Decisions are written down so nobody has to be awake to unblock someone else — and so customers can read the reasoning later.

Documentation first

A feature is designed as a doc page before it is coded: what it measures, how it is defined, what it costs in privacy. The docs site and the glossary are those documents, cleaned up.

A public changelog

Every release is dated and described in plain language, including fixes and the occasional regression. If you want to know whether we are still shipping, that page answers faster than a sales call.

See your data clearly. Find your growth.

Every click, backed by data. Install one line of code and see your first numbers in a minute.

No credit card · Free plans for analytics and chat · Cookieless analytics